Journal of Hepatology 1998; 29: 300–305 Printed in Denmark ¡ All rights reserved Munksgaard ¡ Copenhagen
Copyright C European Association for the Study of the Liver 1998
Journal of Hepatology ISSN 0168-8278
Exposure of hepatic sinusoidal mononuclear cells to UW solution in situ but not ex vivo induces apoptosis Koichi Kinoshita, Iwao Ikai, Takashi Gomi, Michiyuki Kanai, Sigeru Tsuyuki, Tetsuro Hirose, Yasuhiro Kawai, Akira Yamauchi, Takashi Inamoto1, Yukihiro Inomata2, Koichi Tanaka2 and Yoshio Yamaoka Department of Gastroenterological Surgery, 1College of Medical Technology, and 2Department of Transplantation Immunology, Kyoto University Graduate School of Medicine, Kyoto, Japan
Background/Aims: We have previously reported that human hepatic sinusoidal mononuclear cells may have a higher sensitivity to induction of apoptosis than peripheral blood mononuclear cells. In this study, the effects of two different preservation solutions on the functions of those hepatic mononuclear cells were evaluated in living-related liver transplantation. Methods: Ten and 11 liver grafts were perfused via the portal vein with University of Wisconsin solutions (UW group) and Bretschneider’s Histidine-Tryptophan-Ketoglutarate solutions (HTK group), respectively. Hepatic mononuclear cells were isolated from the effluent preservation solution passing through the graft livers. Cytofluorometry, cytotoxic assay, and DNA analysis were performed. Results: There were no significant differences in surface antigens and natural killer activity of hepatic sinusoidal mononuclear cells between the UW and HTK groups. At the time of isolation, the viability
of hepatic sinusoidal mononuclear cells in both groups was more than 99%. In the UW group, the viability of hepatic sinusoidal mononuclear cells decreased to 30% through apoptosis in in vitro culture at 48 h after isolation. In the HTK group, however, their viability was maintained at more than 90% at 48 h in the same culture conditions, and additional exposure to UW solution ex vivo for 30 min did not induce apoptosis. Conclusion: Hepatic sinusoidal mononuclear cells isolated from the UW solution, not from the HTK solution, passing through the liver died through apoptosis, which was not induced by each component of the UW solution, but by exposure in situ.
transplantation, passenger leukocytes are transported to recipients together with the graft (1). These leukocytes are mainly T cells, displaying CD8 predominantly and HLA-DR partially (1,2), and NK cells. Liver sinusoidal passenger leukocytes may be related to the liver graft acceptance in rodent models (3). These unique populations of leukocytes have been suggested to play a pivotal role in the immunogenicity of the transplanted livers (4). Both the University of Wisconsin (UW) solution
and the Bretschneider’s Histidine-Tryptophan-Ketoglutarate (HTK) solution have been reported to be appropriate for clinical use in liver transplantation (5), although some investigators have described their different effects on hepatocytes or hepatic endothelial cells (6–9). Nevertheless, the effect of preservation solutions on passenger leukocytes has not been addressed directly. A major proportion of passenger leukocytes in living-related liver transplantation (LRLT) are the human hepatic sinusoidal mononuclear cells (HSMC), which have been isolated from effluent preservation solution at liver graft harvesting from healthy donors. In this study, to further address the role of the passenger leukocytes in the liver transplantation, we evaluated the effect of two different preservation solutions, UW and HTK solutions, on HSMC of liver grafts in LRLT.
I
Received 8 September 1997; revised 26 February; accepted 27 February 1998
Correspondence: Iwao Ikai, Department of Gastroenterological Surgery, Kyoto University Graduate School of Medicine, 54 Kawahara-cho, Shogoin, Sakyo-ku, Kyoto, 606-8507, Japan. Tel: 81-75-751-3242. Fax: 81-75-751-4263.
300
Key words: Apoptosis; Hepatic sinusoidal mononuclear cells; HTK solution; Living-related liver transplantation; Rejection; University of Wisconsin solution.
UW induces apoptosis of hepatic lymphocytes
Materials and Methods
TABLE 1
Donors and organ procurement Twenty-one donors (nine male and 12 female) for LRLT in Kyoto University Hospital were enrolled in this study. Ten and 11 cases were randomly allocated to the UW and HTK groups, respectively. The study was approved by the Ethics Committee of Kyoto University. There were no significant differences in age, sex, and preoperative examination (GOT, GPT, LDH, HBsAg, and HBsAb) of donors between the two groups (Table 1). Left lateral or left hepatic segmentectomies were performed by the same team according to our standardized method for harvesting the graft in all donors (10,11). Ten liver grafts were perfused with UW solution (UW group) and 11 grafts were perfused with HTK solution (HTK group).
Characteristics of donors
Isolation of mononuclear cells from liver graft All samples were obtained in compliance with institutional guidelines. The effluent preservation solution passing through the graft livers was collected after the harvesting of the liver (2). HSMC were isolated from these effluent solutions after centrifugation on FicollPaque (Pharmacia LKB, Uppsala, Sweden). Cell viability was determined by trypan blue dye exclusion. The HSMC were cultured for 0 to 48 h in RPMI 1640 medium (Sigma Chemical Co., St. Louis, MO, USA) with 10% FCS (Whittaker Bioproducts, Walkersville, MD, USA), at a concentration of 1.0¿106 cells/ml, in 96-well plastic culture plates (InterMed, DK-4000 Roskilde, Denmark) at 37æC, 5% CO2. At the indicated times, aliquots of cells were removed and cell viability was determined. Phenotype analysis by flow cytometry The surface phenotype of lymphocytes was analyzed by single- or two-color immunofluorescence tests. All aliquots of FITC- or phycoerythrin (PE)-conjugated monoclonal antibodies including anti-CD3, anti-CD4, anti-CD8, anti-CD16, anti-CD20, anti-CD25 (anti-IL2 receptor a chain (IL-2Ra)) anti-CD56, anti-TcR-1 (specific for an ab chain framework determinant), and anti-TcRg/d-1 (reacting with a framework epitope of the gd T cell receptor) antibodies were purchased from Becton Dickinson (Mountain View, CA, USA). FITC-conjugated Mik-b1 (anti-IL-2Rb) was obtained from Nichirei Co.Ltd. (Japan). Cells were sorted on FACScan (Becton Dickinson) and analyzed. NK cell activity To evaluate the NK activity of HSMC, 51Cr-labeled K562 cells derived from human erythroleukemia (kindly provided by Dr. M. Maeda, Chest Disease In-
Evaluation criteria
UW group
HTK group
Number of donors Age (year)
10 30.8∫7.5 (23–46)
11 33.6∫5.4 (24–44)
Sex Male Female Laboratory data GOT (IU/l) GPT (IU/l) LDH (IU/l) HBsAg (π) HBsAb (π)
3 7 13.8∫3.9 (9–22) 14.2∫11.0 (2–43) 290∫31 (248–336) 0/10 0/10
6 5 16.1∫4.7 (24–44) 17.5∫9.2 (6–37) 312∫59 (234–429) 0/11 0/11
UWΩUniversity of Wisconsin solution. HTKΩHistidine-tryptophan-ketoglutarate solution. Values are expressed as mean ∫SD.
stitute, Kyoto University, Japan) were used as targets in a 4-h radioisotope release assay, as described elsewhere (12). Briefly, 1¿105 of K562 cells were incubated with 100 mCi of 51Cr sodium chromate (DuPont, Boston, MA, USA) at 37æC for 1 h. One ¿104 of K562 were washed three times with complete medium and added to various numbers of effector cells plated in 96well round-bottom culture plates. After a 4-h incubation, the supernatant of each well was collected using a supernatant collecting system (Dai Nippon Medical Co., Japan), and the radioactivity of the supernatant was counted by a gamma counter (Abbot Laboratories). Specific lysis was calculated by the following formula: % specific lysisΩ(sample releaseªspontaneous release)/(maximal releaseªspontaneous release) ¿100%. Gel analysis of DNA Detection of DNA fragmentation was performed as described previously (13). HSMC were cultured at 1¿107 cells/well in 6-well culture plates. At the indicated time intervals, aliquots of cells were washed and pelleted. DNA was extracted by the phenolchloroform method and analyzed by 1.5% agarose gel electrophoresis, followed by ethidium bromide staining. Data analysis Values were expressed as mean ∫SD. Statistical analysis was performed using Student’s t-test for experimental data. P-values of less than 0.05 were considered significant.
301
K. Kinoshita et al.
Results Cellular composition of HSMC The numbers of HSMC isolated from the effluent preservation solution in the UW and HTK groups were 0.91∫0.21¿106 and 1.24∫0.48¿106 cells/g ¡ wet weight, respectively, which were not significantly different. Viabilities of HSMC in both groups were greater than 99%, by trypan blue exclusion, at the time of isolation. Table 2 shows the phenotypic characteristics of HSMC in both groups. There was no difference in the proportion of CD3π T cells and in the proportion of CD3ª CD56π natural killer cells between the two groups. Nor were CD4π/CD8π ratios, CD25π (IL-2Raπ), IL2Rbπ, TcRabπ, TcRgdπ, HLA-DRπ, and CD3π CD56π proportions in T cells, or the percentage of CD3ª CD56π and CD16ª CD56π NK cells significantly different. CD20π B cells were a minor population of lymphoid cells in both HSMC. NK activity NK cell activity was assessed by a 4-h radioisotope release assay employing 51Cr-labeled K562 cells. HSMC showed strong cytotoxicity against K562 cells at any effector-to-target ratios, with no significant differences between HSMC of both groups (Fig. 1). In all cases, the NK cell activity of sinusoidal MC was always higher than that of peripheral blood mononuclear cells (PBMC), as we previously reported (2). Viability of HSMC Viability was assessed by trypan blue exclusion. In the UW group, appreciable HSMC apoptosis was seen in normal culture conditions as early as 12 h after isolation. Viability of UW/HSMC was reduced to
66.0∫3.9% at 12 h after isolation, 44.9∫8.9% at 24 h, and to 26.8∫9.3% at 48 h. On the other hand, HSMC in the HTK group retained greater than 90% viability for at least 48 h after isolation (p∞0.05) (Fig. 2). Gel analysis of DNA extracted from HSMC Gel analysis of DNA extracted from UW/HSMC cultured for 12, 24, and 48 h after isolation revealed the
Fig. 1. Natural killer (NK) activity of hepatic sinusoidal mononuclear cells (HSMC) of University of Wisconsin solution (UW) group and histidine-tryptophan-ketoglutarate solutions (HTK) group. Specific lysis of K562 target cells after 4 h 51Cr release assay by HSMC of the UW group (solid bar) and HSMC of HTK group (open bar) at various effector/target ratios. Data are expressed as means ∫SD. There were no significant differences in NK activity between the two groups.
TABLE 2 Phenotype of hepatic sinusoidal mononuclear cells (HSMC) (percentage of positive cells) Surface antigen
UW group HTK group (nΩ10) (nΩ11)
CD3π CD4π CD8ª CD4ª CD8π CD4π/CD8π ratio CD3π TcRabπ CD3π TcRgdπ CD3π CD25π CD3πIL-2Rb CD3πHLA-DRπ CD3π CD56π CD3ª CD56π CD16π CD56π CD16ª CD56π CD20π
58.5∫14.2 7.5∫3.9 65.9∫13.4 0.13∫0.11 28.2∫5.7 6.51∫3.4 1.7∫1.0 11.0∫4.1 21.3∫7.7 19.1∫6.2 38.1∫9.8 29.9∫10.7 29.9∫6.1 4.9∫1.7
Values are expressed as mean ∫SD.
302
63.8∫6.0 10.9∫8.4 58.0∫6.2 0.19∫0.15 29.9∫5.8 7.9∫3.9 2.9∫1.2 12.6∫3.9 19.3∫5.7 22.2∫4.9 27.9∫7.5 18.3∫6.5 32.3∫5.7 4.2∫1.2
Fig. 2. Changes in viability of HSMC of UW (open squares) and HTK (closed circles) groups. HSMC of both groups were cultured for 48 h. Viability was measured by trypan blue dye exclusion at the indicated timepoints. Data are expressed as mean percentage viability ∫SD of each group. * p∞0.05.
UW induces apoptosis of hepatic lymphocytes TABLE 3 Effect of additional exposure to UW or HTK solution on HSMC of the HTK group Time (h)
0 24 48
% Viability UW solution
HTK solution
99.0∫0.6 96.7∫1.3 94.1∫2.9
99.8∫0.4 96.2∫1.6 93.3∫2.7
HSMC in five cases of HTK group were cultured for 48 h after exposure to UW or HTK solution for 30 min. Their viabilities were measured by trypan blue dye exclusion tests. Values are expressed as mean ∫SD of five experiments.
TABLE 4
Fig. 3. Gel analysis of DNA from cultured HSMC. DNA samples were extracted from HSMC of the two different donor livers: one was perfused with HTK solution and the other with UW solution. In HSMC isolated from HTK solution, no DNA fragmentation was seen at the time of isolation and at 48 h after culture (lane 1, 2). HSMC isolated in UW solution did not show any fragmentation at the time of isolation (lane 3). However, the characteristic ladder pattern of DNA fragmentation was detected in HSMC at 12 h after isolation (lane 4, 5, and 6 indicate 12, 24, and 48 h, respectively). Molecular sizes in kilobases are indicated on the left for the gel.
Characteristics of recipients Evaluation criteria
UW group
HTK group
Nunber of recipients Blood group matching (ABO) Age (years)
10 identical 4.2∫4.1 (6 months– 10 years)
11 identical 3.3∫3.9 (8 months– 11 years)
Sex Male Female Weight (kg)
Effect of UW solution on HSMC of the HTK group Immediately after isolation, HSMC in the HTK group were exposed to either UW solution or HTK solution for 30 min, which was similar to the period of exposure to UW solution from the graft perfusion to the cell isolation in the UW group. We then cultured cells in RPMI 1640 medium with 10% fetal calf serum and assessed cell viability at the indicated times. The additional exposure in vitro to the UW or HTK solution had no effect on the viability of HSMC in the HTK group (Table 3). Clinical results Table 4 shows the clinical features of the recipients. There were no significant differences in cold, warm, and total ischemic time between the two groups. There were also no statistically significant differences in postoperative laboratory data including transaminases for the week after transplantation. Acute cellular rejection was observed in 4/10 in the UW group and 4/11 in the HTK group within 6 months after transplantation,
6 5
14.1∫9.5 (4.6–31.2) 165∫82 (89–301) 47∫9 (33–63) 216∫81 (127–344)
13.3∫11.2 (5.6–40) 105∫31 (66–178) 52∫16 (31–83) 157∫34 (97–217)
GPT (IU/l) 1st postoperative day 3rd postoperative day 7th postoperative day
155∫65 125∫96 65∫43
180∫132 177∫144 77∫33
Number of rejections (%)
4/10 (40%)
4/11 (36%)
Cold ischemia time (min) Warm ischemia time (min)
DNA laddering characteristic of the DNA fragmentation of apoptosis (Fig. 3). In HTK/HSMC no DNA fragmentation was seen even at 48 h.
3 7
Total ischemia time (min)
Values are expressed as mean ∫SD.
which was not a statistically significant difference between the two groups.
Discussion We have previously reported that HSMC were isolated from the UW solution passing through the graft liver at graft harvesting from normal healthy donors in LRLT (2). These HSMC consisted principally of CD8π T and NK cells, and were found to die in vitro through apoptosis after isolation from the liver, by gel analysis of DNA and morphology (13). This apoptosis was prevented by the addition of reducing agents, such as reduced glutathione or 2-mercaptoethanol, to the culture. We therefore considered that HSMC might have a higher sensitivity than PBMC to extracellular reducing conditions. We showed that HSMC from nor-
303
K. Kinoshita et al.
mal healthy donors expressed some activation markers, with LFA-1 expression higher than PBMC (2,14). In addition, HSMC were shown to express HLA-DR (class 2 antigen), which is also an activation marker of T cells. These phenotypic results are compatible with the report of Pruvot et al. (15). It has been reported that activated lymphocytes are more likely to die through apoptosis than those in a resting state (16,17). HSMC might therefore be more sensitive to an apoptotic stimulus than PBMC. In this study, phenotypic and cytotoxic analysis showed that HSMC in each group consisted of a similar population of lymphoid cells and had a similar expression rate of activation markers. However, HSMC isolated from HTK solution maintained a viability of greater than 90%, although more than 70% of HSMC in the UW group underwent cell death through apoptosis 48 h after isolation in the same culture conditions. These results suggest that the activation status of HSMC is not a main cause of the induction of apoptosis. UW solution differs significantly from HTK solution, having a higher oncotic pressure (15–25 mmHg) and potassium concentration (125 mM) than HTK solution (5). Among several properties of UW solution, the presence of adenosine (18) and hyperosmotic stress (19) were reported to induce apoptosis ex vivo. Oxidative stress has also been reported to induce apoptosis in several types of cells, which may be blocked by antioxidants (20,21). A lower reducing activity of UW solution than of HTK solution might be another factor inducing apoptosis. Therefore, there are several candidates for a trigger of UW-induced apoptosis. In this study, however, additional ex vivo exposure of HSMC in the HTK group to UW solution did not induce apoptosis. This suggests that the factors mentioned above are insufficient to induce apoptosis in HSMC ex vivo. Rather, we suggest that exposure in situ to UW solution by portal perfusion might provoke apoptosis in HSMC. For example, it has been reported that UW solution has a higher viscosity index than HTK solution (4.8 versus 1.8, respectively) (22). Immunohistological examinations of the liver have revealed that HSMC adhere to Kupffer cells or sinusoidal endothelial cells (23). Through these adhesions, HSMC may receive signals related to HSMC apoptosis. There are different perfusion effects on the liver graft between these solutions (6), which conceivably have a different mechanical effect on HSMC on detaching from sinusoidal lining cells. While many reports have shown no significant clinical differences between UW and HTK solutions, some animal studies demonstrated that hepatocellular injuries increased in a extended preservation time-dependent manner (9). In our LRLT program, however,
304
preservation time was at most only 3 or 4 h, and no significant differences in parenchymal cellular damage were shown between these solutions in the early postoperative laboratory data such as GPT. In liver transplantation, there are protean factors related to acute cellular rejection, such as graft condition, time or method of graft preservation and posttransplant immunosuppressive protocols (24). In the present study of 21 ABO-identical LRLT cases, there were negligible deviations in the above-mentioned factors. On the other hand, a major proportion of HSMC was thought to be passenger leukocytes. Passenger leukocytes may have an immunomodulatory role, such as induction of a state of clonal anergy, veto cell effect (25), or liver graft acceptance in rodent models (3). For example, an experimental study suggested that donortype antigens or enriched donor-cell infusions may be tolerogenic stimuli (26). Therefore, if passenger lymphocytes died through apoptosis after perfusion of UW solution, immunological reactions between the cells of donor and recipient might be altered and affect the post-transplantational immunological events. In this study, however, the rates of acute cellular rejection occurring within 6 months after operation were similar in the UW and HTK groups. Finally, human hepatic sinusoidal mononuclear cells died through apoptosis isolated from the liver by perfusion of UW solution, but not of HTK solution. In addition, this apoptosis was induced by exposure to UW solution in situ, but not by each component of the solution. Further studies are required for evaluation of the effects of the preservation solutions on HSMC, not only from various immunological viewpoints including hematopoietic microchimerism (27), but also from clinical observations made over a long postoperative period.
Acknowledgments We would like to thank Dr. Masaaki Awane, Massachusetts General Hospital, Boston, Massachusetts for technical advice, and Drs. Keizo Furuke and Eda T. Bloom, Food and Drug Administration, Bethesda, Maryland, USA for helpful suggestions. This work was supported in part by grants 09557105, 09470266, 09671302, 09307026 from the Scientific Research Fund of the Ministry of Education, Japan, and grant JSPSRFTF96I00204 from the Research for the Future of Japan Society for the Promotion of Science.
References 1. Schlitt HJ, Raddats G, Steinhoff G, Wonigeit K, Pichlmayr R. Passenger lymphocytes in human liver allografts and their potential role after transplantation. Transplantation 1993; 56: 951–5.
UW induces apoptosis of hepatic lymphocytes 2. Kanai M, Inamoto T, Kinoshita K, Yamaoka Y. Phenotypic analysis and natural killer activity of human sinusoidal mononuclear cells. Current Topics in Mucosal Immunology 1993: 253–6. 3. Olszewski WL, Jasklowska-Englisz M, Sadowska-Rytta D, Lukomska B, Maksymowicz M. Liver sinusoidal passenger lymphocytes reveal suppressive properties and prolong allograft survival. Transpl Proc 1995; 27: 194–5. 4. Schlitt HJ, Kanehiro H, Raddats G, Steinhoff G, Richter N, Nashan B, et al. Persistence of donor lymphocytes in liver allograft recipients. Transplantation 1993; 56: 1001–7. 5. Erhard J, Lange R, Scherer R, Kox WJ, Bretschneider HJ, Gebhard MM, et al. Comparison of histidine-tryptophanketoglutarate (HTK) solution versus University of Wisconsin (UW) solution for organ preservation in human liver transplantation. Transpl Int 1994; 7: 177–81. 6. Aminalai A, Kehrer G, Grossmann F, Richter J, Bretschneider HJ. Morphological investigation of the porcine liver directly following preservation with Euro-Collins, University of Wisconsin, and Bretschneider’s HTK solution. Langenbecks Arch Chir 1992; 377: 81–8. 7. Walcher F, Marzi I, Bu¨hren V. Liver preservation in UW and HTK solution: the influence on sinusoidal perfusion after liver transplantation. Transplant Proc 1991; 23: 2376–7. 8. Rauen U, Erhard J, Ku¨hnhenrich P, Lange R, Moissidis M, Eigler FW, et al. Nonparenchymal cell and hepatocellular injury to human liver grafts assessed by enzyme-release into the perfusate. Langenbecks Arch Chir 1994; 379: 241– 7. 9. van Gulik TM, Reinders ME, Nio R, Frederiks WM, Bosma A, Klopper PJ. Preservation of canine liver grafts using HTK solution. Transplantation 1994; 57: 167–71. 10. Yamaoka Y, Ozawa K, Mori K, Morimoto T, Shimahara Y, Zaima M, et al. New device for harvesting a hepatic graft from a living donor. Transplantation 1991; 52: 157–60. 11. Tanaka K,. Uemoto S, Tokunaga Y, Fujita S, Sano K, Yamamoto E, et al. Living related liver transplantation. Transplant Proc 1992; 24: 2252–3. 12. Kanai M, Inamoto T, Kanaya S, Kinoshita K, Fujimura N, Yamauchi A. Inhibition of epidermal growth factor-stimulated hepatocytes proliferation by autologous sinusoidal mononuclear cells in rat liver. Clin Immunol Immunopathol 1993; 69: 23–8. 13. Kinoshita K, Kanai M, Awane M, Furuke K, Shirakata Y, Gomi T, et al. Reducing environment protects sinusoidal lymphocytes isolated from normal human liver from apoptosis. J Hepatol 1997; 26: 103–10.
14. Lanier L, Myle LA, Civin CI, Loken MR, Phillip JH. The relationship of CD16 (LEU-11) and LEU-19 (NKH-1) antigen expression on human peripheral blood NK cells and cytotoxic T lymphocytes. J Immunol 1986; 136: 4480–6. 15. Pruvot FR, Navarro F, Janin A, Labalette M, LecomteHoucke M, Gambiez L, et al. Characterization, quantification, and localization of passenger T lymphocytes and NK cells in human liver before transplantation. Transpl Int 1995; 8: 273–9. 16. Wesselborg S, Janssen O, Kabelitz D. Induction of activationdriven death (apoptosis) in activated but not resting peripheral blood T cells. J Immunol 1993; 150: 4338–45. 17. Lucas M, Solano F, Sanz A. Induction of programmed cell death (apoptosis) in mature lymphocytes. FEBS Lett 1991; 279: 19–20. 18. Tanaka Y, Yoshihara K, Tsuyuki M, Kamiya T. Apoptosis induced by adenosine in human leukemia HL-60 cells. Exp Cell Res 1994; 213: 242–52. 19. Matthews CC, Feldman EL. Insulin-like growth factor I rescues SH-SY5Y human neuroblastoma cells from hyperosmotic induced programmed cell death. J Cell Physiol 1996; 166: 323–31. 20. Kurita T, Namiki H. Apoptotic cell death induced by serum and its prevention by thiols. J Cell Physiol 1994; 161: 63–70. 21. Sandstrom PA, Mannie MD, Buttke TM. Inhibition of activated-induced death in T cell hybridomas by thiol antioxidants: oxidative stress as a mediator of apoptosis. J Leukoc Biol 1994; 55: 221–6. 22. Schilling M, Redaelli C, Friess H, Laeuffer J, Bu¨chler M. Temperature dependence of proton buffering capacity of HTK, Euro-Collins, and UW solution. Transplant Proc 1996; 28: 343–4. 23. Winnock M, Lafon ME, Boulard A, Ferrer AM, Saric J, Dubuisson L, et al. Characterization of liver-associated natural killer cells in patients with liver tumors. Hepatology 1991; 4: 676–82. 24. Inomata Y, Tanaka K, Egawa H, Uemoto S, Ozaki N, Okajima H, et al. The evolution of immunosuppression with FK506 in pediatric living-related liver transplantation. Transplantation 1996; 61: 247–52. 25. Thomas JM, Verbanac KM, Thomas FT. The veto mechanism in transplant tolerance. Transplant Rev 1991; 5: 209–14. 26. Browne BJ, Kahan BD. Transplantation 1994: the year in review. Clin Transpl 1994: 317–40. 27. Elwood ET, Larsen CP, Maurer DH, Routenberg KL, Neylan JF, Whelchel JD, et al. Microchimerism and rejection in clinical transplantation. Lancet 1997; 349: 1358–60.
305